It was not because of the legal issues. FreeBSD 4 was widely considered as the pinnacle of FreeBSD's technial advantage over Linux, for example.
I don't see how this disputes the simple fact that BSD Unix ran into major legal hurdles in the early days of its migration to the x86-based commodity software sphere, thus helping Linux gain an early lead.
BSDs were run on some of the busiest websites on the internet for years. Hotmail, Yahoo, cdrom.
You seem to be trying to say that somehow BSD Unix systems such as FreeBSD were doing great in the early days, then suddenly tanked in quality and were thus surpassed by Linux. The flaw in the argument, of course, is that technical quality has far less to do with popularity than technically minded people like us (I assume you fall into this category, for argument's sake) would like to think. A large number of businesses using BSD Unix early on were just using it -- not further developing it, not distributing it -- and in that respect had little or nothing to do with popularizing the system for anyone else. Meanwhile, the Linux community grew by leaps and bounds because of an initial low barrier to entry and low level of uncertainty over its future that bootstrapped the critical mass needed to create a self-expanding effort. As more users (some of whom became contributors) got involved, the grassroots marketing power of the community grew over time, and soon enough Linux had become enough of a buzzword that it became synonymous with "open source" in the minds of many. It's possible that, when the business explosion of Linux occurred (thus pushing it past BSD Unix significantly in business usage), many of the business adopters did so on the say-so of technical employees who essentially didn't know BSD Unix existed as a viable option.
All of that has little or nothing to do with comparative technical benefits of one system over another. In point of fact, FreeBSD and common Linux-based systems tend to be trade-offs in terms of performance; FreeBSD tends to be more stable for most purposes; Linux-based systems are often much more focused on being "user friendly"; and FreeBSD tends to be much more oriented to low-overhead system administration, consistency of experience, and "correctness" of development practices and system design. Both, in other words, have their advantages. Pretending that "Linux" somehow "won" by being technically better must be a case of either confirmation bias or simple ignorance.
Of course there are categories where GPL licensed code is gaining ground over BSD and similar licensed software.
That's correct. In this regard, libman overstepped himself. Your example of version control systems is pretty much the canonical example, where the top version control systems in the open source world seem to be (in descending order) Git, Mercurial, Bazaar, Fossil, and Subversion. While Subversion still boasts a huge userbase (it may well outstrip Git in that regard), much of that is essentially "legacy" rather than a matter of its popularity. Of the five, the first three were created and distributed under the terms of the GPL (a copyleft license); Fossil uses a BSD license (a copyfree license); and Subversion initially used a copyfree license as well. It gets worse for the copyfree side: when the Apache Software Foundation took over control of the Subversion project, it relicensed it under the terms of the Apache License 2.0, which (while many people dubiously label it "permissive") is not a copyfree license. It's worth noting that the ASF has kind of a center-of-gravity effect on licensing, though; every project it adopts is relicensed to Apache License 2.0, including both the previously copyfree licensed Subversion and the previously copyleft licensed OpenOffice.org.
The most notable thing about the version control example, however, is the fact that it is one of the exceptions to the general trend toward mor
I'm better served by public domain open source, both when I've released software as open source and when I've used others open source. GPL has only ever restricted what I want to do with open source software, not given me more freedom.
You do say GPL has restricted what you want to do.
I think you mean "to quote BasilBrush", and "BasilBrush does say", because it was not me who said that in this thread.
I said "presumably this means you develop commercial software" because the requirements around providing source code tend to allow users to give copies away to others for free which goes against the profit motive.
In other words, you jumped to conclusions.
Apparently that's incorrect - you seem a bit angry about my supposition.
I doubt you are a psychologist, and even if you are you should try to refrain from diagnosing mental state in Internet discussions. I'm not angry about anything. I'm simply pointing out that you have utterly failed to address the core point of BasilBrush's comments.
But if we take that off the table it only leaves a couple options for what you don't like. And I do say "what you don't like" deliberately. At this point the GPL does not directly prevent you from using software, adding to the software, incorporating code into your own software, or redistributing any of that. It does require you to do certain things in some of those case.
Yes, it is true that the GPL requires certain actions that may not be practical or desirable in some cases when "using" covered software, for a very broad definition of "using" that includes stuff like distribution and development among other things. By doing so, it restricts the "use" of the software in cases where such actions are impractical or undesirable.
It requires your combined work to be GPL licensed should you give it to someone (which you may not like).
Worse than that, it may make it effectively impossible to give it to someone else in some circumstances, such as in combined distribution with works covered by other copyleft licenses.
It requires you to offer source code (which may be an inconvenience).
It may be more than an inconvenience. There have been cases where I did not have access to source code myself and thus have not been able to legally give a copy of GPLed works to others, through no fault of anyone higher up the chain of distribution. Consider the case of some upstream distributor who ceased distributing a given version, exhausted the source distribution period, and thus stopped distributing the source as well. I'm then disallowed from distributing the copy I have, by law, due to the fact that I can't even point someone at the upstream distributor's copy of the sources.
Personally I would have accepted my presumption - the diminished potential for commercial use is the strongest and most reasonable objection to the GPL (IMHO of course).
It's an unnecessary, often inaccurate excuse to try to shut down or ignore arguments from others.
The remaining motives seem more like personal preference and laziness.
. . . and here, it starts to look like you have zero interest in actually discussing the benefits and detriments of the GPL and other copyleft licenses, per se, and only really want to denigrate the character of those who prefer other licenses, shut them up, and live in a world where you don't have to think about uncomfortable factors that may lead some to reject copyleft licensing for their purposes. If that's not your intent, you should probably rethink your methods.
And lastly:
I specifically avoid contributing code to anything licensed under copyleft terms because I do not want other people restricting how third p
One major difference is that Linux distributions made it possible for mere mortals to acquire, install, and use a Unixlike OS while BSD was still a superbitch.
When is this "was" of which you speak?
I see no evidence that more cutting-edge development is going on in BSDland, there's tons of stuff in Linuxland that is going on that isn't part of mainline... yet.
How is it my problem that you aren't paying enough attention to notice developments in the BSD Unix world?
Statistically nobody cares that FreeBSD supported amd64 before Linux did because by the time anybody really had hardware to run it on, it was working.
. . . except all the people who cared.
There has been room for a Free/Open PC-BSD community since the sources to BSD-4.3-lite were released. It didn't really come together until Linux broke on the scene. The only comfort you have is being ahead of Minix.
What the heck are you talking about? Are you aware that PC-BSD is the name of a particular distribution of FreeBSD? Are you aware that FreeBSD arrived as a contemporary of Slackware, with both systems being (finally) usable systems? Is there some point you're trying to make that is getting lost in the haphazard way you hand-wave away facts and just make vague pronouncements that are difficult to nail down to having a particular meaning reflective of reality?
I think that, if you're going to make a point about it being okay to have extramarital sex when it's allowed by the spouse, you should be suggesting that the extramarital sex in this case should not be called "adultery", rather than justifying adultery per se.
He has no sense of context or consequence, apparently. If his reasoning is logical, it must also be exclusionary of much of the evidence necessary to make his logic relevant. Logically, software that is open has benefits over software that is not open. Logically, software that is restricted to be "open" by some narrow definition will not be open in other ways -- so there he fails to take into account more than the most simplistic, narrow application of a basic observation, and as a result his "logic" drives him to the point of incorrect conclusions. It's like there's a natural law that goes something like "If only A, then B, but if A plus C, then D," and ignored the word "only" and everything after B when the most common case is A plus C.
Are you incapable of addressing the core argument? Must you make wild guesses about people's motivations and attack those rather than simply addressing what someone actually said?
Do you think I must be a developer of closed source commercial software just because I prefer copyfree licensing over copyleft licensing? I'm not. I have not written a piece of software distributed to others under closed source, strictly enforced copyright in years. I contribute to open source projects, and release programs I have written from scratch only under open source licenses. I specifically avoid contributing code to anything licensed under copyleft terms because I do not want other people restricting how third parties can use my code. Period.
. . . so no, one does not have to be the equivalent of Bill Gates, Steve Jobs, or Larry Ellison to prefer unrestricted licensing. In fact, while I prefer copyfree licenses over the public domain, the atrocious state of copyright law that actually makes dedication to the public domain invalid in many jurisdictions is the big reason for that fact. Otherwise, I'd probably just use public domain dedications for everything.
Unlike RMS, I'm not a hypocrite about my licensing preferences, either. While people might argue about the hypocrisy (or lack thereof) involved in the "freedom" rhetoric surrounding copyleft licensing, my point is not about that. It's about the fact that I apply the same licensing policy to not only software, but basically everything else I create as well. Even my standard email signature comes with a license notice releasing the original content of my emails under a copyfree license. Meanwhile, RMS prefers to use a noncommercial, nonderivative, or noncommercial-nonderivative license for anything he creates that is not software. Screw that.
It's better to be fanatic for freedom than passively allowing its destruction.
It's better to actually advance the cause of freedom than to eventually become a detriment to it, and a parody of yourself in the process, while claiming to advance the cause of freedom.
I think you missed the point people make when they quote RMS' comments on Jobs. The point, I think, is not to say that RMS is bad for saying negative things about Jobs; it's to point out that if it's okay for RMS to say negative (but true) things about Jobs when he died, it's surely just as okay for people in comments here to say negative (but true) things about RMS when he hasn't even died (yet?).
Don't give people shit for expressing that view, calling it indecent as if they went over to Stallman's house (if he has one -- I don't actually know) and physically injected some virus into him to cause his illness.
Hardly. The Linux project and the open source BSD Unix project were neck-and-neck in early development. Linus Torvalds just managed to attract a larger crowd for initial interest, while the BSD Unix project was being managed a bit less publicly in an attempt to ensure everything was well-structured when a complete system was unveiled. FreeBSD and NetBSD were actually released the same year as Slackware, the oldest surviving Linux distribution; 386BSD whence FreeBSD and NetBSD were forked was released in 1992, only a few months after the initial release of a not-yet-really-usable Linux kernel in late 1991, and that slight difference in release dates made all the difference in terms of early popularity on which Linux built continuing later popularity.
The various major BSD Unix systems have exceeded Linux systems in terms of "modernity" in many cases along the way, in fact, including the originally very notable -- but now largely forgotten -- fact that FreeBSD had support for the AMD64 architecture well ahead of the Linux kernel. Meanwhile, Ubuntu keeps getting press for growing popularity and dramatic changes in desktop defaults, while PC-BSD advances the state of the art of desktop system management including software management tools, installer capabilities, and other things significantly more important than background image galleries, the corner of the screen used for the menu button, and the default width of the scrollbar.
Linux-based systems, FreeBSD, NetBSD, and OpenBSD all contribute significantly to leaps forward in functionality and presentation, and only someone who is too narrowly focused on Linux-based systems is likely to come to the honest conclusion that Linux is unequivocally, generally more "modern" at any point in its history than the major BSD Unix competitors.
They had problems with the GPL. They just figured that the problems with switching before that point added up to enough to somewhat eclipse the problems with GPLv2, at least enough that the resources they felt they would be justified in devoting to developing a replacement for GCC would not be sufficient to make much progress. Apple, however, did some serious work to that end, and Clang reaching a point of usable maturity closely coincided with the increasing difficulties involved in avoiding the GPLv3 versions of GCC, such that suddenly various projects are abandoning GCC in favor of LLVM/Clang, and others are considering other alternatives than the GCC as well. At least, that's how it looked to me.
A more reasonable approach would be to say that to the extent a person's ability to make informed decisions is hindered by lack of mature mental, emotional, and neurological development, that person should not be regarded as being meaningfully capable of consent in cases of decisions particularly prone to traumatic consequences when decisions are made poorly. This would, for instance, exempt a typical six year old child from being legally capable of independently consenting to sexual intercourse.
I have no idea whether RMS would agree or disagree with that approach to things, of course, because so far as I am aware he has not made any remarks specifically about the case of pederasty. If he would disagree, and if he would think that young children are capable of sexual consent, he might have intentionally avoided making an argument on the subject because of a rare case of realizing something he's going to say might not play well in public. If he would agree, or have an even more strict view of how to disallow children from legally consenting to sexual intercourse, he might have simply regarded the answer as so obvious as to not need elaboration.
If I had to guess on the subject of child pornography, though, I might venture to suggest he probably wouldn't object to wholly fictional illustrations of a pedophilic nature, even though he might well object to photographic images of similar character. That's a guess, though, inferring from statements that do not actually bear directly on that question.
I'm not so sure that's why. The sunk cost fallacy is the thinking that leads people to believe that when they've invested a lot in something, they lose that investment if they switch to some alternative, and as a result count that lost investment as part of the cost of switching. The reality, of course, is that the investment is already lost, and all that should actually be counted is direct cost of switching plus the difference in return on investment going forward. While an argument could be made that there's a substantial community emotional investment in RMS as a figurehead, and as such a sunk cost fallacy may be in play in the way people refuse to acknowledge that there may be reasons to cease treating RMS as the patron saint of open source ("free") software, I think it is not a very strong argument.
The stronger argument is, I believe, community ego-identification with the rationale behind early and continuing attachment to Stallman. The difference is that ego-identification with a choice is the case where people feel their self-image and reputation are vulnerable to the perception that they make bad choices if they allow a particular decision to be questioned and its rationale undermined. In this case, people who have tightly coupled their open source software related belief systems to RMS and his conception of Free Software have doggedly defended their decision to do so until they now cannot admit they are wrong without suffering a severe blow to their egos. As such, the cognitive dissonance that arises between continued attachment as a means of trying to prove themselves right and the very practical and reasonable pressures to abandon that attachment in favor of more productive ideals and methodologies for achieving their nominal goals is mitigated by compartmentalizing their thoughts, disallowing any meaningful integration of their views on RMS and the FSF on one hand with their observations of the actual effects of related policies in the real world.
In short, they refuse to think too deeply about the consequences of the choices they've made to avoid having to admit that they made bad choices, and as a result they compound the severity of the problem by continuing to faithfully support those bad decisions.
All this being said, gcc is still the most popular C compiler today.
Its popularity is trending downward, and with good reason. Its popularity relative to other open source alternatives is largely an accident of history, much like that of Linux. In fact, GCC's popularity has almost entirely been built on the popularity of Linux. Meanwhile, I'm using Clang/LLVM for all of my C development, and to do all software installation and updating on my FreeBSD systems.
Even if he didn't personally write every single line of code of the GNU software corpus, he's still the guy who kickstarted it all, who managed that huge project, and who kept everything together. Even though a boat needs rowers, it still need a helmsman too.
I haven't checked the list lately, but from what I recall there were basically only three significant software systems in the GNU project -- the rest is a collection of simple tools that people reimplement for fun, for projects in their Bachelor level university projects, and so on. Those three things are GCC and its accessories, the kernel, and Emacs -- the latter of which one might rightly regard as not even a part of GNU proper, because it's not a reasonable piece of a core OS. The kernel in the only popular OS family particularly dependent on the GNU project was created entirely outside of the GNU project; I speak of the Linux kernel, in fact. That leaves GCC, whose niche could easily have been filled by tools like PCC, TenDRA, and other old C compilers, if not for an accident of history that gave GCC a relatively early popularity boost.
Even worse, GNU project software tends to be characterized by ugly design, lack of clean extensibility, featuritis, and unnecessary incompatibility with standards and de facto standard ways of doing things, as well as an exclusionary attitude amongst core developers such that it is very difficult for outsiders to contribute to development. This is why things like tmux, Clang, and other projects intended to replace GNU tools are so popular once undertaken. Only the laziness virtue of programmers -- don't reimplement what's already there if you don't need to -- has allowed the GNU project to remain relevant as long as it has.
Without a wealth of basic user-space tools to put on top of the kernel, Linux might have a slower start than it had.
The Linux project could easily have gotten those tools somewhere else. The GNU tools just basically fell in its lap.
Well, speaking for FreeBSD, we use Clang/LLVM and are in the process of removing GCC. We're hoping to flip the switch to defaulting to Clang this weekend. OpenBSD is in the process of switching to PCC. NetBSD supports multiple compilers.
Excellent! Speaking as a FreeBSD user (and port maintainer), this makes me very happy. I've been using Clang/LLVM as my standard compiler on FreeBSD installs for a while now, and am quite pleased to hear that the project is already on the cusp of eliminating my need to make a config change to ensure that Clang is what my system uses to compile ports and base system software.
Are you a committer working on this? If so, you have my thanks.
RMS has spent his life fighting for your rights. And when he gets sick all you can do is pile dirt on him?
It's pretty obvious that some people disagree with RMS about what constitutes a "right" (e.g. forcing other people to distribute what they possess under some circumstances). You're probably going to have to get someone to agree that what Stallman advocates is good before you'll get some people to agree that Stallman has spent his life fighting for something good.
In short, the fact you and RMS declared RMS to be some kind of selfless crusader for the betterment of the world does not translate to a compelling argument that he is, in fact, such a person. Those who view him as more harmful than good and who believe many of his views are at best misguided and more likely downright unethical or evil are not piling dirt on someone who spent his life fighting for their rights, in their view. If you want to debate whether RMS has been doing good, you'll have to actually debate whether he has been doing good, instead of just declaring it and trying to use that to shame someone for disagreeing with you.
If there's anything in what you said with which I wouldn't agree, it's the fact it looks like you're implying that we can't really do significantly better than SQL. While we haven't yet developed something significantly better (in all important ways), as far as I'm aware, I really believe we could. I think it's mostly a combination of inertia, orthodoxy, and compatibility issues that stops us.
(Who would then have to call the them back a couple days later? That strikes me as more creepy than the preggo-score.)
Maybe the manager asked for the telephone number when the guy came to complain so that he could call back a couple days later and offer them some kind of conciliatory special deal at the store (like discounts on something). On the other hand, maybe the manager was trying to arrange for the guy's family to no longer get (at the time, presumed faulty) targeted advertising, and was calling back to give them an update on the process (once again having explicitly asked for contact information for just this purpose).
I don't know if it was actually creepy. We don't know enough details to come to a conclusion about that, I think.
Slashdot Mirror
I don't see how this disputes the simple fact that BSD Unix ran into major legal hurdles in the early days of its migration to the x86-based commodity software sphere, thus helping Linux gain an early lead.
You seem to be trying to say that somehow BSD Unix systems such as FreeBSD were doing great in the early days, then suddenly tanked in quality and were thus surpassed by Linux. The flaw in the argument, of course, is that technical quality has far less to do with popularity than technically minded people like us (I assume you fall into this category, for argument's sake) would like to think. A large number of businesses using BSD Unix early on were just using it -- not further developing it, not distributing it -- and in that respect had little or nothing to do with popularizing the system for anyone else. Meanwhile, the Linux community grew by leaps and bounds because of an initial low barrier to entry and low level of uncertainty over its future that bootstrapped the critical mass needed to create a self-expanding effort. As more users (some of whom became contributors) got involved, the grassroots marketing power of the community grew over time, and soon enough Linux had become enough of a buzzword that it became synonymous with "open source" in the minds of many. It's possible that, when the business explosion of Linux occurred (thus pushing it past BSD Unix significantly in business usage), many of the business adopters did so on the say-so of technical employees who essentially didn't know BSD Unix existed as a viable option.
All of that has little or nothing to do with comparative technical benefits of one system over another. In point of fact, FreeBSD and common Linux-based systems tend to be trade-offs in terms of performance; FreeBSD tends to be more stable for most purposes; Linux-based systems are often much more focused on being "user friendly"; and FreeBSD tends to be much more oriented to low-overhead system administration, consistency of experience, and "correctness" of development practices and system design. Both, in other words, have their advantages. Pretending that "Linux" somehow "won" by being technically better must be a case of either confirmation bias or simple ignorance.
That's correct. In this regard, libman overstepped himself. Your example of version control systems is pretty much the canonical example, where the top version control systems in the open source world seem to be (in descending order) Git, Mercurial, Bazaar, Fossil, and Subversion. While Subversion still boasts a huge userbase (it may well outstrip Git in that regard), much of that is essentially "legacy" rather than a matter of its popularity. Of the five, the first three were created and distributed under the terms of the GPL (a copyleft license); Fossil uses a BSD license (a copyfree license); and Subversion initially used a copyfree license as well. It gets worse for the copyfree side: when the Apache Software Foundation took over control of the Subversion project, it relicensed it under the terms of the Apache License 2.0, which (while many people dubiously label it "permissive") is not a copyfree license. It's worth noting that the ASF has kind of a center-of-gravity effect on licensing, though; every project it adopts is relicensed to Apache License 2.0, including both the previously copyfree licensed Subversion and the previously copyleft licensed OpenOffice.org.
The most notable thing about the version control example, however, is the fact that it is one of the exceptions to the general trend toward mor
To quote you:
You do say GPL has restricted what you want to do.
I think you mean "to quote BasilBrush", and "BasilBrush does say", because it was not me who said that in this thread.
I said "presumably this means you develop commercial software" because the requirements around providing source code tend to allow users to give copies away to others for free which goes against the profit motive.
In other words, you jumped to conclusions.
Apparently that's incorrect - you seem a bit angry about my supposition.
I doubt you are a psychologist, and even if you are you should try to refrain from diagnosing mental state in Internet discussions. I'm not angry about anything. I'm simply pointing out that you have utterly failed to address the core point of BasilBrush's comments.
But if we take that off the table it only leaves a couple options for what you don't like. And I do say "what you don't like" deliberately. At this point the GPL does not directly prevent you from using software, adding to the software, incorporating code into your own software, or redistributing any of that. It does require you to do certain things in some of those case.
Yes, it is true that the GPL requires certain actions that may not be practical or desirable in some cases when "using" covered software, for a very broad definition of "using" that includes stuff like distribution and development among other things. By doing so, it restricts the "use" of the software in cases where such actions are impractical or undesirable.
It requires your combined work to be GPL licensed should you give it to someone (which you may not like).
Worse than that, it may make it effectively impossible to give it to someone else in some circumstances, such as in combined distribution with works covered by other copyleft licenses.
It requires you to offer source code (which may be an inconvenience).
It may be more than an inconvenience. There have been cases where I did not have access to source code myself and thus have not been able to legally give a copy of GPLed works to others, through no fault of anyone higher up the chain of distribution. Consider the case of some upstream distributor who ceased distributing a given version, exhausted the source distribution period, and thus stopped distributing the source as well. I'm then disallowed from distributing the copy I have, by law, due to the fact that I can't even point someone at the upstream distributor's copy of the sources.
Personally I would have accepted my presumption - the diminished potential for commercial use is the strongest and most reasonable objection to the GPL (IMHO of course).
It's an unnecessary, often inaccurate excuse to try to shut down or ignore arguments from others.
The remaining motives seem more like personal preference and laziness.
. . . and here, it starts to look like you have zero interest in actually discussing the benefits and detriments of the GPL and other copyleft licenses, per se, and only really want to denigrate the character of those who prefer other licenses, shut them up, and live in a world where you don't have to think about uncomfortable factors that may lead some to reject copyleft licensing for their purposes. If that's not your intent, you should probably rethink your methods.
And lastly:
Alas, the real world does not much seem to match up with how you describe it.
Thanks for your contributions, then, and thank you for the updates. In case you were ever unsure, users like me really appreciate your work.
One major difference is that Linux distributions made it possible for mere mortals to acquire, install, and use a Unixlike OS while BSD was still a superbitch.
When is this "was" of which you speak?
I see no evidence that more cutting-edge development is going on in BSDland, there's tons of stuff in Linuxland that is going on that isn't part of mainline... yet.
How is it my problem that you aren't paying enough attention to notice developments in the BSD Unix world?
Statistically nobody cares that FreeBSD supported amd64 before Linux did because by the time anybody really had hardware to run it on, it was working.
. . . except all the people who cared.
There has been room for a Free/Open PC-BSD community since the sources to BSD-4.3-lite were released. It didn't really come together until Linux broke on the scene. The only comfort you have is being ahead of Minix.
What the heck are you talking about? Are you aware that PC-BSD is the name of a particular distribution of FreeBSD? Are you aware that FreeBSD arrived as a contemporary of Slackware, with both systems being (finally) usable systems? Is there some point you're trying to make that is getting lost in the haphazard way you hand-wave away facts and just make vague pronouncements that are difficult to nail down to having a particular meaning reflective of reality?
I think that, if you're going to make a point about it being okay to have extramarital sex when it's allowed by the spouse, you should be suggesting that the extramarital sex in this case should not be called "adultery", rather than justifying adultery per se.
I really hope that was sarcasm.
He has no sense of context or consequence, apparently. If his reasoning is logical, it must also be exclusionary of much of the evidence necessary to make his logic relevant. Logically, software that is open has benefits over software that is not open. Logically, software that is restricted to be "open" by some narrow definition will not be open in other ways -- so there he fails to take into account more than the most simplistic, narrow application of a basic observation, and as a result his "logic" drives him to the point of incorrect conclusions. It's like there's a natural law that goes something like "If only A, then B, but if A plus C, then D," and ignored the word "only" and everything after B when the most common case is A plus C.
The only part of what this operating system does worth a damn is editing text -- and that part belongs to viper mode.
ignoring all the relevant commentary
Oh, that must mean SuperBanana is wrong! Right? Right . . . ?
Hmm. Wait a minute. No, that's not what that means at all. Darn.
You presumably develop commercial software.
Are you incapable of addressing the core argument? Must you make wild guesses about people's motivations and attack those rather than simply addressing what someone actually said?
Do you think I must be a developer of closed source commercial software just because I prefer copyfree licensing over copyleft licensing? I'm not. I have not written a piece of software distributed to others under closed source, strictly enforced copyright in years. I contribute to open source projects, and release programs I have written from scratch only under open source licenses. I specifically avoid contributing code to anything licensed under copyleft terms because I do not want other people restricting how third parties can use my code. Period.
. . . so no, one does not have to be the equivalent of Bill Gates, Steve Jobs, or Larry Ellison to prefer unrestricted licensing. In fact, while I prefer copyfree licenses over the public domain, the atrocious state of copyright law that actually makes dedication to the public domain invalid in many jurisdictions is the big reason for that fact. Otherwise, I'd probably just use public domain dedications for everything.
Unlike RMS, I'm not a hypocrite about my licensing preferences, either. While people might argue about the hypocrisy (or lack thereof) involved in the "freedom" rhetoric surrounding copyleft licensing, my point is not about that. It's about the fact that I apply the same licensing policy to not only software, but basically everything else I create as well. Even my standard email signature comes with a license notice releasing the original content of my emails under a copyfree license. Meanwhile, RMS prefers to use a noncommercial, nonderivative, or noncommercial-nonderivative license for anything he creates that is not software. Screw that.
It's better to be fanatic for freedom than passively allowing its destruction.
It's better to actually advance the cause of freedom than to eventually become a detriment to it, and a parody of yourself in the process, while claiming to advance the cause of freedom.
I think you missed the point people make when they quote RMS' comments on Jobs. The point, I think, is not to say that RMS is bad for saying negative things about Jobs; it's to point out that if it's okay for RMS to say negative (but true) things about Jobs when he died, it's surely just as okay for people in comments here to say negative (but true) things about RMS when he hasn't even died (yet?).
The point seemed pretty clear to me:
Don't give people shit for expressing that view, calling it indecent as if they went over to Stallman's house (if he has one -- I don't actually know) and physically injected some virus into him to cause his illness.
Shamed into modernity . . . ?
Hardly. The Linux project and the open source BSD Unix project were neck-and-neck in early development. Linus Torvalds just managed to attract a larger crowd for initial interest, while the BSD Unix project was being managed a bit less publicly in an attempt to ensure everything was well-structured when a complete system was unveiled. FreeBSD and NetBSD were actually released the same year as Slackware, the oldest surviving Linux distribution; 386BSD whence FreeBSD and NetBSD were forked was released in 1992, only a few months after the initial release of a not-yet-really-usable Linux kernel in late 1991, and that slight difference in release dates made all the difference in terms of early popularity on which Linux built continuing later popularity.
The various major BSD Unix systems have exceeded Linux systems in terms of "modernity" in many cases along the way, in fact, including the originally very notable -- but now largely forgotten -- fact that FreeBSD had support for the AMD64 architecture well ahead of the Linux kernel. Meanwhile, Ubuntu keeps getting press for growing popularity and dramatic changes in desktop defaults, while PC-BSD advances the state of the art of desktop system management including software management tools, installer capabilities, and other things significantly more important than background image galleries, the corner of the screen used for the menu button, and the default width of the scrollbar.
Linux-based systems, FreeBSD, NetBSD, and OpenBSD all contribute significantly to leaps forward in functionality and presentation, and only someone who is too narrowly focused on Linux-based systems is likely to come to the honest conclusion that Linux is unequivocally, generally more "modern" at any point in its history than the major BSD Unix competitors.
They had problems with the GPL. They just figured that the problems with switching before that point added up to enough to somewhat eclipse the problems with GPLv2, at least enough that the resources they felt they would be justified in devoting to developing a replacement for GCC would not be sufficient to make much progress. Apple, however, did some serious work to that end, and Clang reaching a point of usable maturity closely coincided with the increasing difficulties involved in avoiding the GPLv3 versions of GCC, such that suddenly various projects are abandoning GCC in favor of LLVM/Clang, and others are considering other alternatives than the GCC as well. At least, that's how it looked to me.
A more reasonable approach would be to say that to the extent a person's ability to make informed decisions is hindered by lack of mature mental, emotional, and neurological development, that person should not be regarded as being meaningfully capable of consent in cases of decisions particularly prone to traumatic consequences when decisions are made poorly. This would, for instance, exempt a typical six year old child from being legally capable of independently consenting to sexual intercourse.
I have no idea whether RMS would agree or disagree with that approach to things, of course, because so far as I am aware he has not made any remarks specifically about the case of pederasty. If he would disagree, and if he would think that young children are capable of sexual consent, he might have intentionally avoided making an argument on the subject because of a rare case of realizing something he's going to say might not play well in public. If he would agree, or have an even more strict view of how to disallow children from legally consenting to sexual intercourse, he might have simply regarded the answer as so obvious as to not need elaboration.
If I had to guess on the subject of child pornography, though, I might venture to suggest he probably wouldn't object to wholly fictional illustrations of a pedophilic nature, even though he might well object to photographic images of similar character. That's a guess, though, inferring from statements that do not actually bear directly on that question.
I'm not so sure that's why. The sunk cost fallacy is the thinking that leads people to believe that when they've invested a lot in something, they lose that investment if they switch to some alternative, and as a result count that lost investment as part of the cost of switching. The reality, of course, is that the investment is already lost, and all that should actually be counted is direct cost of switching plus the difference in return on investment going forward. While an argument could be made that there's a substantial community emotional investment in RMS as a figurehead, and as such a sunk cost fallacy may be in play in the way people refuse to acknowledge that there may be reasons to cease treating RMS as the patron saint of open source ("free") software, I think it is not a very strong argument.
The stronger argument is, I believe, community ego-identification with the rationale behind early and continuing attachment to Stallman. The difference is that ego-identification with a choice is the case where people feel their self-image and reputation are vulnerable to the perception that they make bad choices if they allow a particular decision to be questioned and its rationale undermined. In this case, people who have tightly coupled their open source software related belief systems to RMS and his conception of Free Software have doggedly defended their decision to do so until they now cannot admit they are wrong without suffering a severe blow to their egos. As such, the cognitive dissonance that arises between continued attachment as a means of trying to prove themselves right and the very practical and reasonable pressures to abandon that attachment in favor of more productive ideals and methodologies for achieving their nominal goals is mitigated by compartmentalizing their thoughts, disallowing any meaningful integration of their views on RMS and the FSF on one hand with their observations of the actual effects of related policies in the real world.
In short, they refuse to think too deeply about the consequences of the choices they've made to avoid having to admit that they made bad choices, and as a result they compound the severity of the problem by continuing to faithfully support those bad decisions.
All this being said, gcc is still the most popular C compiler today.
Its popularity is trending downward, and with good reason. Its popularity relative to other open source alternatives is largely an accident of history, much like that of Linux. In fact, GCC's popularity has almost entirely been built on the popularity of Linux. Meanwhile, I'm using Clang/LLVM for all of my C development, and to do all software installation and updating on my FreeBSD systems.
Even if he didn't personally write every single line of code of the GNU software corpus, he's still the guy who kickstarted it all, who managed that huge project, and who kept everything together. Even though a boat needs rowers, it still need a helmsman too.
I haven't checked the list lately, but from what I recall there were basically only three significant software systems in the GNU project -- the rest is a collection of simple tools that people reimplement for fun, for projects in their Bachelor level university projects, and so on. Those three things are GCC and its accessories, the kernel, and Emacs -- the latter of which one might rightly regard as not even a part of GNU proper, because it's not a reasonable piece of a core OS. The kernel in the only popular OS family particularly dependent on the GNU project was created entirely outside of the GNU project; I speak of the Linux kernel, in fact. That leaves GCC, whose niche could easily have been filled by tools like PCC, TenDRA, and other old C compilers, if not for an accident of history that gave GCC a relatively early popularity boost.
Even worse, GNU project software tends to be characterized by ugly design, lack of clean extensibility, featuritis, and unnecessary incompatibility with standards and de facto standard ways of doing things, as well as an exclusionary attitude amongst core developers such that it is very difficult for outsiders to contribute to development. This is why things like tmux, Clang, and other projects intended to replace GNU tools are so popular once undertaken. Only the laziness virtue of programmers -- don't reimplement what's already there if you don't need to -- has allowed the GNU project to remain relevant as long as it has.
Without a wealth of basic user-space tools to put on top of the kernel, Linux might have a slower start than it had.
The Linux project could easily have gotten those tools somewhere else. The GNU tools just basically fell in its lap.
Well, speaking for FreeBSD, we use Clang/LLVM and are in the process of removing GCC. We're hoping to flip the switch to defaulting to Clang this weekend. OpenBSD is in the process of switching to PCC. NetBSD supports multiple compilers.
Excellent! Speaking as a FreeBSD user (and port maintainer), this makes me very happy. I've been using Clang/LLVM as my standard compiler on FreeBSD installs for a while now, and am quite pleased to hear that the project is already on the cusp of eliminating my need to make a config change to ensure that Clang is what my system uses to compile ports and base system software.
Are you a committer working on this? If so, you have my thanks.
RMS has spent his life fighting for your rights. And when he gets sick all you can do is pile dirt on him?
It's pretty obvious that some people disagree with RMS about what constitutes a "right" (e.g. forcing other people to distribute what they possess under some circumstances). You're probably going to have to get someone to agree that what Stallman advocates is good before you'll get some people to agree that Stallman has spent his life fighting for something good. In short, the fact you and RMS declared RMS to be some kind of selfless crusader for the betterment of the world does not translate to a compelling argument that he is, in fact, such a person. Those who view him as more harmful than good and who believe many of his views are at best misguided and more likely downright unethical or evil are not piling dirt on someone who spent his life fighting for their rights, in their view. If you want to debate whether RMS has been doing good, you'll have to actually debate whether he has been doing good, instead of just declaring it and trying to use that to shame someone for disagreeing with you.
If there's anything in what you said with which I wouldn't agree, it's the fact it looks like you're implying that we can't really do significantly better than SQL. While we haven't yet developed something significantly better (in all important ways), as far as I'm aware, I really believe we could. I think it's mostly a combination of inertia, orthodoxy, and compatibility issues that stops us.
You say this as though you are disagreeing with me, which seems odd.
(Who would then have to call the them back a couple days later? That strikes me as more creepy than the preggo-score.)
Maybe the manager asked for the telephone number when the guy came to complain so that he could call back a couple days later and offer them some kind of conciliatory special deal at the store (like discounts on something). On the other hand, maybe the manager was trying to arrange for the guy's family to no longer get (at the time, presumed faulty) targeted advertising, and was calling back to give them an update on the process (once again having explicitly asked for contact information for just this purpose). I don't know if it was actually creepy. We don't know enough details to come to a conclusion about that, I think.
-- "What Are They Gonna Do When Were All Using Freenet"
Statistical profiling, of course.